Update generated Python Op docs.

Change: 147308976

9 files changed, 544 insertions, 2 deletions

diff --git a/tensorflow/g3doc/api_docs/python/contrib.linalg.md b/tensorflow/g3doc/api_docs/python/contrib.linalg.md
index bb78cd4434..2eb11e6766 100644
--- a/tensorflow/g3doc/api_docs/python/contrib.linalg.md
+++ b/tensorflow/g3doc/api_docs/python/contrib.linalg.md
@@ -280,6 +280,39 @@ Determinant for every batch member.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperator.diag_part(name='diag_part')` {#LinearOperator.diag_part}
+
+Efficiently get the [batch] diagonal part of this operator.
+
+If this operator has shape `[B1,...,Bb, M, N]`, this returns a
+`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
+`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
+
+```
+my_operator = LinearOperatorDiag([1., 2.])
+
+# Efficiently get the diagonal
+my_operator.diag_part()
+==> [1., 2.]
+
+# Equivalent, but inefficient method
+tf.matrix_diag_part(my_operator.to_dense())
+==> [1., 2.]
+```
+
+##### Args:
+
+
+*  <b>`name`</b>: A name for this `Op`.
+
+##### Returns:
+
+
+*  <b>`diag_part`</b>: A `Tensor` of same `dtype` as self.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperator.domain_dimension` {#LinearOperator.domain_dimension}
 
 Dimension (in the sense of vector spaces) of the domain of this operator.
@@ -787,6 +820,39 @@ Determinant for every batch member.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperatorDiag.diag_part(name='diag_part')` {#LinearOperatorDiag.diag_part}
+
+Efficiently get the [batch] diagonal part of this operator.
+
+If this operator has shape `[B1,...,Bb, M, N]`, this returns a
+`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
+`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
+
+```
+my_operator = LinearOperatorDiag([1., 2.])
+
+# Efficiently get the diagonal
+my_operator.diag_part()
+==> [1., 2.]
+
+# Equivalent, but inefficient method
+tf.matrix_diag_part(my_operator.to_dense())
+==> [1., 2.]
+```
+
+##### Args:
+
+
+*  <b>`name`</b>: A name for this `Op`.
+
+##### Returns:
+
+
+*  <b>`diag_part`</b>: A `Tensor` of same `dtype` as self.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperatorDiag.domain_dimension` {#LinearOperatorDiag.domain_dimension}
 
 Dimension (in the sense of vector spaces) of the domain of this operator.
@@ -1321,6 +1387,39 @@ Determinant for every batch member.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperatorIdentity.diag_part(name='diag_part')` {#LinearOperatorIdentity.diag_part}
+
+Efficiently get the [batch] diagonal part of this operator.
+
+If this operator has shape `[B1,...,Bb, M, N]`, this returns a
+`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
+`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
+
+```
+my_operator = LinearOperatorDiag([1., 2.])
+
+# Efficiently get the diagonal
+my_operator.diag_part()
+==> [1., 2.]
+
+# Equivalent, but inefficient method
+tf.matrix_diag_part(my_operator.to_dense())
+==> [1., 2.]
+```
+
+##### Args:
+
+
+*  <b>`name`</b>: A name for this `Op`.
+
+##### Returns:
+
+
+*  <b>`diag_part`</b>: A `Tensor` of same `dtype` as self.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperatorIdentity.domain_dimension` {#LinearOperatorIdentity.domain_dimension}
 
 Dimension (in the sense of vector spaces) of the domain of this operator.
@@ -1829,6 +1928,39 @@ Determinant for every batch member.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperatorScaledIdentity.diag_part(name='diag_part')` {#LinearOperatorScaledIdentity.diag_part}
+
+Efficiently get the [batch] diagonal part of this operator.
+
+If this operator has shape `[B1,...,Bb, M, N]`, this returns a
+`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
+`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
+
+```
+my_operator = LinearOperatorDiag([1., 2.])
+
+# Efficiently get the diagonal
+my_operator.diag_part()
+==> [1., 2.]
+
+# Equivalent, but inefficient method
+tf.matrix_diag_part(my_operator.to_dense())
+==> [1., 2.]
+```
+
+##### Args:
+
+
+*  <b>`name`</b>: A name for this `Op`.
+
+##### Returns:
+
+
+*  <b>`diag_part`</b>: A `Tensor` of same `dtype` as self.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperatorScaledIdentity.domain_dimension` {#LinearOperatorScaledIdentity.domain_dimension}
 
 Dimension (in the sense of vector spaces) of the domain of this operator.
@@ -2327,6 +2459,39 @@ Determinant for every batch member.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperatorMatrix.diag_part(name='diag_part')` {#LinearOperatorMatrix.diag_part}
+
+Efficiently get the [batch] diagonal part of this operator.
+
+If this operator has shape `[B1,...,Bb, M, N]`, this returns a
+`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
+`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
+
+```
+my_operator = LinearOperatorDiag([1., 2.])
+
+# Efficiently get the diagonal
+my_operator.diag_part()
+==> [1., 2.]
+
+# Equivalent, but inefficient method
+tf.matrix_diag_part(my_operator.to_dense())
+==> [1., 2.]
+```
+
+##### Args:
+
+
+*  <b>`name`</b>: A name for this `Op`.
+
+##### Returns:
+
+
+*  <b>`diag_part`</b>: A `Tensor` of same `dtype` as self.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperatorMatrix.domain_dimension` {#LinearOperatorMatrix.domain_dimension}
 
 Dimension (in the sense of vector spaces) of the domain of this operator.
@@ -2820,6 +2985,39 @@ Determinant for every batch member.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperatorTriL.diag_part(name='diag_part')` {#LinearOperatorTriL.diag_part}
+
+Efficiently get the [batch] diagonal part of this operator.
+
+If this operator has shape `[B1,...,Bb, M, N]`, this returns a
+`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
+`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
+
+```
+my_operator = LinearOperatorDiag([1., 2.])
+
+# Efficiently get the diagonal
+my_operator.diag_part()
+==> [1., 2.]
+
+# Equivalent, but inefficient method
+tf.matrix_diag_part(my_operator.to_dense())
+==> [1., 2.]
+```
+
+##### Args:
+
+
+*  <b>`name`</b>: A name for this `Op`.
+
+##### Returns:
+
+
+*  <b>`diag_part`</b>: A `Tensor` of same `dtype` as self.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperatorTriL.domain_dimension` {#LinearOperatorTriL.domain_dimension}
 
 Dimension (in the sense of vector spaces) of the domain of this operator.
@@ -3357,7 +3555,7 @@ Determinant for every batch member.
 
 - - -
 
-#### `tf.contrib.linalg.LinearOperatorUDVHUpdate.diag` {#LinearOperatorUDVHUpdate.diag}
+#### `tf.contrib.linalg.LinearOperatorUDVHUpdate.diag_arg` {#LinearOperatorUDVHUpdate.diag_arg}
 
 If this operator is `A = L + U D V^H`, this is the diagonal of `D`.
 
@@ -3371,6 +3569,39 @@ If this operator is `A = L + U D V^H`, this is `D`.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperatorUDVHUpdate.diag_part(name='diag_part')` {#LinearOperatorUDVHUpdate.diag_part}
+
+Efficiently get the [batch] diagonal part of this operator.
+
+If this operator has shape `[B1,...,Bb, M, N]`, this returns a
+`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
+`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
+
+```
+my_operator = LinearOperatorDiag([1., 2.])
+
+# Efficiently get the diagonal
+my_operator.diag_part()
+==> [1., 2.]
+
+# Equivalent, but inefficient method
+tf.matrix_diag_part(my_operator.to_dense())
+==> [1., 2.]
+```
+
+##### Args:
+
+
+*  <b>`name`</b>: A name for this `Op`.
+
+##### Returns:
+
+
+*  <b>`diag_part`</b>: A `Tensor` of same `dtype` as self.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperatorUDVHUpdate.domain_dimension` {#LinearOperatorUDVHUpdate.domain_dimension}
 
 Dimension (in the sense of vector spaces) of the domain of this operator.
@@ -3420,6 +3651,13 @@ List of graph dependencies of this `LinearOperator`.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperatorUDVHUpdate.is_diag_positive` {#LinearOperatorUDVHUpdate.is_diag_positive}
+
+If this operator is `A = L + U D V^H`, this hints `D > 0` elementwise.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperatorUDVHUpdate.is_non_singular` {#LinearOperatorUDVHUpdate.is_non_singular}
 
 
@@ -3889,6 +4127,39 @@ Determinant for every batch member.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperatorComposition.diag_part(name='diag_part')` {#LinearOperatorComposition.diag_part}
+
+Efficiently get the [batch] diagonal part of this operator.
+
+If this operator has shape `[B1,...,Bb, M, N]`, this returns a
+`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
+`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
+
+```
+my_operator = LinearOperatorDiag([1., 2.])
+
+# Efficiently get the diagonal
+my_operator.diag_part()
+==> [1., 2.]
+
+# Equivalent, but inefficient method
+tf.matrix_diag_part(my_operator.to_dense())
+==> [1., 2.]
+```
+
+##### Args:
+
+
+*  <b>`name`</b>: A name for this `Op`.
+
+##### Returns:
+
+
+*  <b>`diag_part`</b>: A `Tensor` of same `dtype` as self.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperatorComposition.domain_dimension` {#LinearOperatorComposition.domain_dimension}
 
 Dimension (in the sense of vector spaces) of the domain of this operator.
diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.contrib.linalg.LinearOperatorDiag.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.contrib.linalg.LinearOperatorDiag.md
index 46ad4a2e8d..f4796c04c1 100644
--- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.contrib.linalg.LinearOperatorDiag.md
+++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.contrib.linalg.LinearOperatorDiag.md
@@ -237,6 +237,39 @@ Determinant for every batch member.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperatorDiag.diag_part(name='diag_part')` {#LinearOperatorDiag.diag_part}
+
+Efficiently get the [batch] diagonal part of this operator.
+
+If this operator has shape `[B1,...,Bb, M, N]`, this returns a
+`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
+`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
+
+```
+my_operator = LinearOperatorDiag([1., 2.])
+
+# Efficiently get the diagonal
+my_operator.diag_part()
+==> [1., 2.]
+
+# Equivalent, but inefficient method
+tf.matrix_diag_part(my_operator.to_dense())
+==> [1., 2.]
+```
+
+##### Args:
+
+
+*  <b>`name`</b>: A name for this `Op`.
+
+##### Returns:
+
+
+*  <b>`diag_part`</b>: A `Tensor` of same `dtype` as self.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperatorDiag.domain_dimension` {#LinearOperatorDiag.domain_dimension}
 
 Dimension (in the sense of vector spaces) of the domain of this operator.
diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.contrib.linalg.LinearOperatorComposition.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.contrib.linalg.LinearOperatorComposition.md
index 340c9f8cba..a6654ab014 100644
--- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.contrib.linalg.LinearOperatorComposition.md
+++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.contrib.linalg.LinearOperatorComposition.md
@@ -234,6 +234,39 @@ Determinant for every batch member.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperatorComposition.diag_part(name='diag_part')` {#LinearOperatorComposition.diag_part}
+
+Efficiently get the [batch] diagonal part of this operator.
+
+If this operator has shape `[B1,...,Bb, M, N]`, this returns a
+`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
+`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
+
+```
+my_operator = LinearOperatorDiag([1., 2.])
+
+# Efficiently get the diagonal
+my_operator.diag_part()
+==> [1., 2.]
+
+# Equivalent, but inefficient method
+tf.matrix_diag_part(my_operator.to_dense())
+==> [1., 2.]
+```
+
+##### Args:
+
+
+*  <b>`name`</b>: A name for this `Op`.
+
+##### Returns:
+
+
+*  <b>`diag_part`</b>: A `Tensor` of same `dtype` as self.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperatorComposition.domain_dimension` {#LinearOperatorComposition.domain_dimension}
 
 Dimension (in the sense of vector spaces) of the domain of this operator.
diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.contrib.linalg.LinearOperatorIdentity.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.contrib.linalg.LinearOperatorIdentity.md
index 30998f8f9f..80f6b13b73 100644
--- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.contrib.linalg.LinearOperatorIdentity.md
+++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.contrib.linalg.LinearOperatorIdentity.md
@@ -267,6 +267,39 @@ Determinant for every batch member.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperatorIdentity.diag_part(name='diag_part')` {#LinearOperatorIdentity.diag_part}
+
+Efficiently get the [batch] diagonal part of this operator.
+
+If this operator has shape `[B1,...,Bb, M, N]`, this returns a
+`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
+`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
+
+```
+my_operator = LinearOperatorDiag([1., 2.])
+
+# Efficiently get the diagonal
+my_operator.diag_part()
+==> [1., 2.]
+
+# Equivalent, but inefficient method
+tf.matrix_diag_part(my_operator.to_dense())
+==> [1., 2.]
+```
+
+##### Args:
+
+
+*  <b>`name`</b>: A name for this `Op`.
+
+##### Returns:
+
+
+*  <b>`diag_part`</b>: A `Tensor` of same `dtype` as self.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperatorIdentity.domain_dimension` {#LinearOperatorIdentity.domain_dimension}
 
 Dimension (in the sense of vector spaces) of the domain of this operator.
diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.contrib.linalg.LinearOperatorScaledIdentity.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.contrib.linalg.LinearOperatorScaledIdentity.md
index 62511fe5f0..4dc1f88b1f 100644
--- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.contrib.linalg.LinearOperatorScaledIdentity.md
+++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.contrib.linalg.LinearOperatorScaledIdentity.md
@@ -241,6 +241,39 @@ Determinant for every batch member.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperatorScaledIdentity.diag_part(name='diag_part')` {#LinearOperatorScaledIdentity.diag_part}
+
+Efficiently get the [batch] diagonal part of this operator.
+
+If this operator has shape `[B1,...,Bb, M, N]`, this returns a
+`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
+`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
+
+```
+my_operator = LinearOperatorDiag([1., 2.])
+
+# Efficiently get the diagonal
+my_operator.diag_part()
+==> [1., 2.]
+
+# Equivalent, but inefficient method
+tf.matrix_diag_part(my_operator.to_dense())
+==> [1., 2.]
+```
+
+##### Args:
+
+
+*  <b>`name`</b>: A name for this `Op`.
+
+##### Returns:
+
+
+*  <b>`diag_part`</b>: A `Tensor` of same `dtype` as self.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperatorScaledIdentity.domain_dimension` {#LinearOperatorScaledIdentity.domain_dimension}
 
 Dimension (in the sense of vector spaces) of the domain of this operator.
diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.contrib.linalg.LinearOperatorUDVHUpdate.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.contrib.linalg.LinearOperatorUDVHUpdate.md
index 9a6f57b795..6705f62ac6 100644
--- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.contrib.linalg.LinearOperatorUDVHUpdate.md
+++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard3/tf.contrib.linalg.LinearOperatorUDVHUpdate.md
@@ -270,7 +270,7 @@ Determinant for every batch member.
 
 - - -
 
-#### `tf.contrib.linalg.LinearOperatorUDVHUpdate.diag` {#LinearOperatorUDVHUpdate.diag}
+#### `tf.contrib.linalg.LinearOperatorUDVHUpdate.diag_arg` {#LinearOperatorUDVHUpdate.diag_arg}
 
 If this operator is `A = L + U D V^H`, this is the diagonal of `D`.
 
@@ -284,6 +284,39 @@ If this operator is `A = L + U D V^H`, this is `D`.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperatorUDVHUpdate.diag_part(name='diag_part')` {#LinearOperatorUDVHUpdate.diag_part}
+
+Efficiently get the [batch] diagonal part of this operator.
+
+If this operator has shape `[B1,...,Bb, M, N]`, this returns a
+`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
+`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
+
+```
+my_operator = LinearOperatorDiag([1., 2.])
+
+# Efficiently get the diagonal
+my_operator.diag_part()
+==> [1., 2.]
+
+# Equivalent, but inefficient method
+tf.matrix_diag_part(my_operator.to_dense())
+==> [1., 2.]
+```
+
+##### Args:
+
+
+*  <b>`name`</b>: A name for this `Op`.
+
+##### Returns:
+
+
+*  <b>`diag_part`</b>: A `Tensor` of same `dtype` as self.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperatorUDVHUpdate.domain_dimension` {#LinearOperatorUDVHUpdate.domain_dimension}
 
 Dimension (in the sense of vector spaces) of the domain of this operator.
@@ -333,6 +366,13 @@ List of graph dependencies of this `LinearOperator`.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperatorUDVHUpdate.is_diag_positive` {#LinearOperatorUDVHUpdate.is_diag_positive}
+
+If this operator is `A = L + U D V^H`, this hints `D > 0` elementwise.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperatorUDVHUpdate.is_non_singular` {#LinearOperatorUDVHUpdate.is_non_singular}
 
 
diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.contrib.linalg.LinearOperatorMatrix.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.contrib.linalg.LinearOperatorMatrix.md
index ec2653a33c..0f07a4457d 100644
--- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.contrib.linalg.LinearOperatorMatrix.md
+++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.contrib.linalg.LinearOperatorMatrix.md
@@ -224,6 +224,39 @@ Determinant for every batch member.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperatorMatrix.diag_part(name='diag_part')` {#LinearOperatorMatrix.diag_part}
+
+Efficiently get the [batch] diagonal part of this operator.
+
+If this operator has shape `[B1,...,Bb, M, N]`, this returns a
+`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
+`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
+
+```
+my_operator = LinearOperatorDiag([1., 2.])
+
+# Efficiently get the diagonal
+my_operator.diag_part()
+==> [1., 2.]
+
+# Equivalent, but inefficient method
+tf.matrix_diag_part(my_operator.to_dense())
+==> [1., 2.]
+```
+
+##### Args:
+
+
+*  <b>`name`</b>: A name for this `Op`.
+
+##### Returns:
+
+
+*  <b>`diag_part`</b>: A `Tensor` of same `dtype` as self.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperatorMatrix.domain_dimension` {#LinearOperatorMatrix.domain_dimension}
 
 Dimension (in the sense of vector spaces) of the domain of this operator.
diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.contrib.linalg.LinearOperatorTriL.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.contrib.linalg.LinearOperatorTriL.md
index 761aa79c30..403e5dac69 100644
--- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.contrib.linalg.LinearOperatorTriL.md
+++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.contrib.linalg.LinearOperatorTriL.md
@@ -226,6 +226,39 @@ Determinant for every batch member.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperatorTriL.diag_part(name='diag_part')` {#LinearOperatorTriL.diag_part}
+
+Efficiently get the [batch] diagonal part of this operator.
+
+If this operator has shape `[B1,...,Bb, M, N]`, this returns a
+`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
+`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
+
+```
+my_operator = LinearOperatorDiag([1., 2.])
+
+# Efficiently get the diagonal
+my_operator.diag_part()
+==> [1., 2.]
+
+# Equivalent, but inefficient method
+tf.matrix_diag_part(my_operator.to_dense())
+==> [1., 2.]
+```
+
+##### Args:
+
+
+*  <b>`name`</b>: A name for this `Op`.
+
+##### Returns:
+
+
+*  <b>`diag_part`</b>: A `Tensor` of same `dtype` as self.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperatorTriL.domain_dimension` {#LinearOperatorTriL.domain_dimension}
 
 Dimension (in the sense of vector spaces) of the domain of this operator.
diff --git a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.contrib.linalg.LinearOperator.md b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.contrib.linalg.LinearOperator.md
index 00c35baf3c..1d9ba477a7 100644
--- a/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.contrib.linalg.LinearOperator.md
+++ b/tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.contrib.linalg.LinearOperator.md
@@ -258,6 +258,39 @@ Determinant for every batch member.
 
 - - -
 
+#### `tf.contrib.linalg.LinearOperator.diag_part(name='diag_part')` {#LinearOperator.diag_part}
+
+Efficiently get the [batch] diagonal part of this operator.
+
+If this operator has shape `[B1,...,Bb, M, N]`, this returns a
+`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
+`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
+
+```
+my_operator = LinearOperatorDiag([1., 2.])
+
+# Efficiently get the diagonal
+my_operator.diag_part()
+==> [1., 2.]
+
+# Equivalent, but inefficient method
+tf.matrix_diag_part(my_operator.to_dense())
+==> [1., 2.]
+```
+
+##### Args:
+
+
+*  <b>`name`</b>: A name for this `Op`.
+
+##### Returns:
+
+
+*  <b>`diag_part`</b>: A `Tensor` of same `dtype` as self.
+
+
+- - -
+
 #### `tf.contrib.linalg.LinearOperator.domain_dimension` {#LinearOperator.domain_dimension}
 
 Dimension (in the sense of vector spaces) of the domain of this operator.